System combination and score normalization for spoken term detection

Spoken content in languages of emerging importance needs to be searchable to provide access to the underlying information. In this paper, we investigate the problem of extending data fusion methodologies from Information Retrieval for Spoken Term Detection on low-resource languages in the framework...

Full description

Saved in:
Bibliographic Details
Main Authors: Mamou, Jonathan, Jia Cui, Xiaodong Cui, Gales, Mark J. F., Kingsbury, Brian, Knill, Kate, Mangu, Lidia, Nolden, David, Picheny, Michael, Ramabhadran, Bhuvana, Schluter, Ralf, Sethy, Abhinav, Woodland, Philip C.
Format: Conference Proceeding
Language:English
Subjects:
data fusion
Data integration
Hidden Markov models
Indexes
keyword search
Lattices
score normalization
Speech
spoken term detection
system combination
Training
Tuning
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Spoken content in languages of emerging importance needs to be searchable to provide access to the underlying information. In this paper, we investigate the problem of extending data fusion methodologies from Information Retrieval for Spoken Term Detection on low-resource languages in the framework of the IARPA Babel program. We describe a number of alternative methods improving keyword search performance. We apply these methods to Cantonese, a language that presents some new issues in terms of reduced resources and shorter query lengths. First, we show score normalization methodology that improves in average by 20% keyword search performance. Second, we show that properly combining the outputs of diverse ASR systems performs 14% better than the best normalized ASR system.
ISSN:1520-6149
2379-190X
DOI:10.1109/ICASSP.2013.6639278